This article was downloaded by: [Case Western Reserve University]On: 31 October 2014, At: 11:59Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Vertebrate PaleontologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/ujvp20

Eremotherium laurillardi: the panamerican latePleistocene megatheriid slothCástor Cartelle a & Gerardo De Iuliis b ca Instituto de Geociências, Universidade Federal de Minas Gerais , Avenida AntonioCarlos, 31.270, Belo Horizonte, Minas Gerais, Brazilb Department of Zoology , University of Toronto , 25 Harbord Street, Toronto,Ontario, Canada , M5S 1A1c Royal Ontario Museum , 100 Queen's Park, Toronto, Ontario, Canada , M5S 2C6Published online: 24 Aug 2010.

To cite this article: Cástor Cartelle & Gerardo De Iuliis (1995) Eremotherium laurillardi: thepanamerican late Pleistocene megatheriid sloth, Journal of Vertebrate Paleontology, 15:4, 830-841, DOI:10.1080/02724634.1995.10011265

To link to this article: http://dx.doi.org/10.1080/02724634.1995.10011265

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”)contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensorsmake no representations or warranties whatsoever as to the accuracy, completeness, or suitabilityfor any purpose of the Content. Any opinions and views expressed in this publication are the opinionsand views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy ofthe Content should not be relied upon and should be independently verified with primary sources ofinformation. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands,costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly orindirectly in connection with, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial orsystematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution inany form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Journal of Vertebrate Paleontology 15(4):830-841, December 1995© 1995 by the Society of Vertebrate Paleontology

EREMOTHERIUM LAURILLARDI: THE PANAMERICAN LATE PLEISTOCENEMEGATHERIIDSLOTH

CASTOR CARTELLEI and GERARDO DE IULIIS2'Instituto de Geociencias, Universidade Federal de Minas Gerais, Avenida Antonio Carlos,

31.270, Belo Horizonte, Minas Gerais, Brazil;2Department of Zoology, University of Toronto, 25 Harbord Street ,

Toronto, Ontario, Canada, M5S lAl andRoyal Ontario Museum, 100 Queen's Park, Toronto, Ontario, Canada M5S 2C6

ABSTRACT-Remains of Eremotherium, representing a large-sized megatheriid ground sloth, areknown from localities in North, Central, and South America. Usually these remains ,are ~urrently

assigned to the following three species, based largely on geographic provenance: E. laurillardi (Lund),E. mirabile (Leidy), and E. rusconii (Schaub). However, two large, recently ,recovered collections ofEremotherium remains from Jacobina, Bahia, Brazil , and Daytona Beach, Florida, USA, do n?t supportthe separation of these species. Instead, these collections demonstrate the existence of a ,smgle Pan ­american species . The range of variation is larger than was suspected and the morphological charac-teristics used in species distinction are not diagnostically valid. . .. .

The valid name for this species is E. laurillardi (Lund, 1842). The type IS a juvenile molanform(ZMUC 1130) from the Pleistocene of Lagoa Santa, Minas Gerais, Brazil. E. mirabile (Leidy, 1855)and E. rusconii (Schaub, 1935) fall as junior synonyms.

INTRODUCTION

Eremotherium comprises a group ofsmall and large,primarily intertropical Pleistocene megatheriine groundsloths that are known from South America, CentralAmerica and North America. Remains of the largerspecies are assigned currently to Eremotherium laur­illardi (Lund, 1842), E. mirabile (Leidy, 1855), and E .rusconii (Schaub, 1935). Those of a possibly smallerspecies have been described under E . elenense (~off­

stetter, 1949), but its diagnosis is not firmly established(De luliis and Cartelle, 1994), and this report does notresolve the status of th is species.

Eremotherium may be distinguished morphologi­cally from Megatherium primarily through differencesin the skull , molariforms, and manus. In Eremother­ium the premaxillae are triangular, small, and looselyarticulated to each other and to the maxillae, whereasin Megatherium the quadrangular premaxillae are fusedfirmly to each other and to the maxillae, and help forma stout and elongated rostrum. The zygomatic arch andorbit lie more ventrally in Eremotherium, and the ven­tral bulge of the mandible is less pronounced in Ere­motherium than in Megatherium. The mandibularsymphysis extends posteriorly approximately to m I inEremotherium, and to m2 in Megatherium.

The molariforms of Eremotherium and Megather­ium differ in that the pulp cavity is relatively shorterin the former comprising approximately half the ap­icobasal length, whereas in Megatherium the cavityoccupies approximately the basal three-fourths. H.ow­ever, these differences apparently become mamfestduring ontogeny, such that molariforms of youngerjuveniles cannot be consistently distinguished. Fur-

830

ther the transverse crests and valleys of the molari­forms, particularly of the more mesial ones, tend to bemore obliquely oriented in Eremotherium.

The manus of Eremotherium described in the lit­erature (excluding that allocated to E . elenense) retainsdigits III-V, ofwhich digits III and IV bear large, well­formed unguals; only metacarpals (MCs) I and II, fusedwith the trapezium and trapezoid into the Metacarpal­Carpal Complex (MCq, represent the first two digits(Fig. I). The manus of Megatherium possesses digitsII-V, of which digits II-IV bear large, well-formedunguals; the trapezium and MC I form the MCC. Therelationships among the carpal and metacarpal ele­ments are more fully elaborated elsewhere (De luliisand Cartelle, 1994).

Two large collections ofEremotherium remains haverecently been made. One constitutes remains from Toeadas Oncas, Jacobina, Bahia, Brazil, and consists ofapproximately 4,000 skeletal elements (M.N.I. = 36),which are housed at MCL (abbreviations are givenbelow). The second collection, from Daytona BeachBonebed, Daytona Beach, Florida, USA, consists ofapproximately 1,300 elements (M.N.I. = II). Most ofthe latter collection is housed at ROM, but a compos­ite nearly complete skeleton is mounted at DMAS .These collections demonstrate the existence ofa singlePanamerican species of Eremotherium during the latePleistocene. The valid name for this species is E. laur­illardi (Lund, 1842).

MATERIALS AND METHODS

Specimens and localities used in statistical analysesare listed in Appendix I. Other material examined is

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

CARTELLE AND DE IULIIS-PANAMERICAN GROUND SLOTH 831

lun

III A

ps

10cm

sc

mag

mcc

III

lun cun

IVB

v

FIGURE 1. Dorsal (A) and palmar (B) views of the right manus of Eremotherium laurillardi (MeL 9487) . Abbreviations:eun, cuneiform; lun, lunar; mag, magnum; me III, metacarpal III; me IV, metacarpal IV; me V, metacarpal V; mee, metacarpal­carpal complex; p, pisiform; ps, palmar sesamoid; une, unciform; se, scaphoid. The position of ps is uncertain.

too numerous to list , and includes specimens fromother localities in Brazil and the United States, andEcuador, Honduras, Panama, Peru, and Venezuela. Theappropriate institution (see text) may be contacted forspecimen lists. All measurements are in em, and weretaken with large-size calipers, resembling tree-calipers,constructed by staff of the Department of Zoology,University of Toronto. Statistical analyses were per­formed using SAS.

Abbreviations: AMNH, American Museum ofNat­ural History, New York; ANSP, Academy of NaturalSciences, Philadelphia; BMNH, British Museum ofNatural History, London; DMAS, Daytona BeachMuseum ofArts and Sciences, Daytona Beach; FMNH,Field Museum of Natural History, Chicago; MCL,Museu de Ciencias Naturais da Pontificia Univer­sidade Catolica de Minas Gerais, Belo Horizonte;MNRJ, Museu Nacional do Rio de Janeiro, Rio deJaneiro; ROM, Royal Ontario Museum, Toronto; UF,Florida Museum ofNatural History, Gainesville; UFA,Universidade Federal do Acre, Rio Branco; USNM,National Museum ofNatural History, Smithsonian In­stitution, Washington; ZMUC, Zoologisk MuseumUniversitat Copenhagen, Copenhagen.

TAXONOMIC HISTORY

Spillmann (1948) erected Eremotherium carolinensefor late Pleistocene megatheriine remains from Ecua­dor. Hoffstetter (1949), without knowledge of Spill­mann's work, referred material from the same localityas that which had yielded Spillmann's specimens toMegatherium rusconii Schaub, 1935, to which Schaubhad allocated megatheriine remains from Venezuela.Hoffstetter believed, however, that the material re­ferred to M. rusconii was generically distinct from Me­gatherium and erected the genus Schaubia. Hoffstetter(1950) discovered that Schaubia was occupied and ren­amed the genus Schaubitherium. Hoffstetter (1952)subsequently realized that his Ecuadorian material (S.rusconii) was generically identical to Spillmann's Er­emotherium carolinense and he revised his allocation;however, Hoffstetter maintained the validity ofSchaub's E. rusconii primarily on the basis of geo­graphic separation.

During the past forty years numerous species havebeen proposed for Eremotherium remains, a circum­stance largely due to the sparse and often poorly pre­served nature of the material, but also to the lack of

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

832 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 15, NO .4, 1995

consideration for either inter- or intraspecific variationof metric or morphological characters. A third reasonis that researchers had little opportunity to visit col­lections in countries in South America. In addition tothe species formally proposed for Eremotherium areothers originally referred to Megatherium, one (seeMones, 1973) to Scelidotherium, and one (see Ray,1979) to Chelonia, a marine turtle; these have beenreidentified recently as Eremotherium. A list of thenames that have appeared in the literature in chro­nological order follows:

Eremotherium laurillardi (Lund, 1842)E. couperi (Harlan, 1842)E . mirabile (Leidy, 1855)E. quanajuatense (Duges, 1882)E. rusconii (Schaub, 1935)E. hudsoni (White, 1941)E. larensis (Nectario-Maria, 1941)E . earolinense Spillmann, 1948E . elenense (Hoffstetter, 1949)E. venezuelensis (Osten, 1951)E . lundi Paula Couto, 1954E . robustum Porta, 1961E. eueutense Porta, 1961

Paula Couto (1950) stated that the material mountedat MNRJ represented Megatherium laurillardi, whichhe considered a possibly valid name for Brazilianmegatheriine remains. Hoffstetter (1954) decided thatthe type material of laurillardi and the material dis­cussed by Paula Couto belonged in the genus Ere­motherium; he considered E. laurillardi as a possiblyval id species , though based on poor material.

When Paula Couto (1954) revised the South Amer­ican Pleistocene megatheriids he recognized five spe­cies of Eremotherium: E. earolinense, E. rusconii, E .mirabile, E . elenense, and the new species E. lundi. Healso proposed the new subgenus Pseudoeremotheriumfor the material of E . lundi, which he considered to berestricted to Brazil. However, the type material of E.lundi was largely that which Hoffstetter (1954) hadrecognized as E. laurillardi; if the latter is consideredvalid, E . lundi would be its junior synonym. Both au­thors supported the validity ofE . rusconiias the Ven­ezuelan species . Later, Paula Couto (1978) consideredE. earolinense, E . elenense, and E . lundi as synonymsof E . rusconii. Paula Couto (1979) considered E. laur­illardi as insufficiently defined and, following Hoffs­tetter (1949, 1952), recognized E. elenense as a smalleremothere species from South America. Paula Couto(1954) also considered "Megatherium larensis" (sic)Nectario-Maria, 1941 and "M. venezuelensis" (sic) Os­ten , 1951 to be either synonyms ofE . rusconii or nom­ina nuda.

Gazin (1957) was the first to recognize the possibilitythat a single panamerican Eremotherium species ex­isted, based on comparisons of newly discovered Pan­amanian material. He considered E . mirabile to be thevalid name for the possible late Pleistocene panamer-

ican species . He allowed, however, for the possibilitythat distinct South and North American species exist­ed, in which case he considered E. rusconii and E.mirabile, respectively, as valid names. He felt, as hadHoffstetter (1952), that E. laurillardi should have beenignored or synonymized with Megatherium american­um. Also, Gazin (1957) agreed with Paula Couto's(1954) reallocation of M . mirabile to E. mirabile forNorth American remains. Paula Couto (1954) alludedto M. hudsoni and he (Paulo Couto, 1979) listed E.mirabile and E. hudsoni as North American taxa. E.hudsoni was erected by White (1941) on an incompleteungual ofdigit III , possibly ofthe manus, from Florida;it is thus a poorly established name and is best con­sidered a synonym of E. laurillardi. It was mistakenlyjudged to be a Pliocene species because of its occur­rence in the Bone Valley District of Florida; while itis true that in this region Tertiary sediments predom­inate, Pleistocene sediments are also quite commonand have subsequently produced other examples ofEremotherium (S. D. Webb, 1991, pers . comm.), asdiscussed above.

Bocquentin (1979) synonymized E. eueutense, "M.venezuelensis" (sic) and "M. larensis" (sic) with E.rusconii. Cartelle and Boh6rquez (1982) synonymizedE. earolinense with E. rusconii, and E. quanajuatenseand E. elenense with E . mirabile. The synonymy of E.elenense with E. mirabile is considered here as incor­rect, and will be considered more fully elsewhere. E.robustum is based on remains from Fusagasuga, Cun­dinamarca, Colombia, originally assigned by BUrgi(1957) to Megatherium sp. The material includes apoorly preserved postcranium and a nearly completedentary. We agree with de Porta's (1961) allocation toEremotherium, but his description and measurementsdo not support the erection of a new species, as theattributes ascribed to E . robustum fall within the rangeof variation observed in the new eremothere samplesfrom Daytona Beach and Jacobina. We therefore syn­onymize E. robustum with E. laurillardi, as explainedbelow.

Cartelle and Boh6rquez (1982) considered E. laur­illardi to be the valid name for Brazilian eremothereremains, following Paula Couto (1950) and Hoffstetter(1954). Currently, three names are usually recognizedin the literature for Eremotherium: E. laurillardi fromBrazil; E . mirabile from the United States; and E. rus­conii from Colombia (see de Porta, 1961), El Salvador(see Stirton and Gealey, 1949), Ecuador (see Hoffstet­ter, 1949, 1952; Edmund, 1965), Honduras (specimensin FMNH), Mexico (see Polaco-Ramos, 1981), Pan­ama (see Gazin, 1957), Peru (specimen in ROM), andVenezuela (see Bocquentin, 1979; also specimens inAMNH).

COMPARISONS

The many specific names erected for eremothere re­mains during the past forty years were not based oncareful comparison with established taxa, but appar-

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

CARTELLE AND DE IULIIS-PANAMERICAN GROUND SLOTH 833

A

BFIGURE 2. Fibular view ofthe astragalus ofEremotheriumlaurillardi (ROM 22008) demonstrating measurements usedin statistical analyses. A = proximodistal height ; B = anter­oposterior length.

ently on a rather arbitrary geographical basis. Thus,for example, specimens from Brazil have been referredusually to E . laurillardi. This practice has resulted, asmight be expected, in the unjustifiable use of nearlyany quantitative or morphologic character found inexceedingly small samples to justify specific designa­tion. Such diagnoses do not bear close scrutiny in thelight of ordinary consideration of variation. For ex­ample, Cartelle and Bohorquez (1982) cite a differentarrangement ofcarpals as possibly separating Brazilianeremotheres from Gazin's (1957) Panamanian mate­rial , stating that the unciform and centrale are fusedin their material, while the centrale is free in Gazin'ssloth. However, the occurrence of an unfused centralein the Panamanian specimen appears to be an anomaly(see below). Cartelle and Bohorquez cited possible fu­sion of the trapezoid and the magnum, and of thetrapezium with MCs I and II, but figures of the Bra­zilian manus (Cartelle and Bohorquez, 1982:fig. 4b)are nearly identical in this respect to Gazin's (1957:350-351; fig. 2) description and figure for the Pana­manian manus.

Two composite skeletons of E. mirabile from Pan­ama are mounted for exhibition in the USNM, eachwith reconstruction. The larger of these, mounted bi­pedally (USNM 20872), is the one for which Gazin(1957) described the manus. Neither manus may beexamined in detail because the mount is fragile, thespaces between the skeletal elements are plastered, and

the palmar sesamoid obscures the manus in palmarview. The centrale of the left manus appears to be real.However, it is very similar to the medial portion of anormal unciform, and may represent an anomalouscondition in which the medial portion of the unciformis separate from the main body. The right manus couldnot be examined in detail because it is posed in a raisedposition, but the centrale appears to be a plaster re­construction. The smaller mount (USNM 20867) isposed quadrupedally. A centrale is absent, and the un­ciforms, which probably are not from the same indi­vidual, are normal. No other centralia are known toexist. A left unciform in the collection (USNM FieldNumber 82-51) is normal. It appears, then, that thepresence of a centrale is an abnormal variation and isknown only from USNM 20872. Unfortunately, Gazin(1957) chose to describe only the anomalous condition.

A second misconception among workers is that er­emothere species may be distinguished on the basis ofsize. Most studies using size to justify specific sepa­ration focus on the supposed discrepancy between thesize of the South American eremothere remains andthat published by Leidy (1855) for the North AmericanE. mirabile. Since the specimen examined by Leidywas not a particularly large one , workers have generallyassumed, following Ameghino (1889) , that the NorthAmerican eremotheres were smaller than their SouthAmerican counterparts. Preliminary statistical analy­ses of the large collections from Toea das Oncas andDaytona Beach do not support these assumptions.Rather, the large Toea das Oncas sample demonstratesvariation ofnearly 35% in linear measurements amongboth adult and juvenile members of a single popula­tion. Kruskal-Wallis tests for the height and length ofthe astragalus (Fig. 2) of adults, among the more com­monly recovered elements, indicate that no significantdifferences occur among five samples (Fig. 3; see Ap­pendix 1 for samples). Further, the Toea Das Oncassample provides evidence, predominantly from longbones, for striking sexual dimorphism, as anticipatedby Cartelle and Bohorquez (1982). Dimorphism is sug­gested in that skeletal elements from near both endsof the size range may belong to adult and juvenileindividuals. These new data clearly contravene the pre­vious view that Eremotherium species could be dis­tinguished on the basis of size.

Similarly, the remains from Toea das Oncas andDaytona Beach display an extensive variation in shapeand form of nearly all skeletal elements and their ar­ticular facets , indicating that the small morphologicalvariations reported in the literature do not constitutevalid diagnoses of Eremotherium species. Indeed, allreported differences ma y be observed in both the Toeadas Oncas and Daytona Beach collections, except forGazin's (1957) report ofa centrale in one Panamanianspecimen, as discussed above. In summary, all sup­posedly diagnostic and quantitative features of latePleistocene eremotheres are subsumed in the new, larg­er samples from Jacobina and Daytona Beach; thus,

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

834 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 15, NO .4, 1995

15 ....._.,....-~-...,..-.....,-.....,~-8

5

3

1&

43

SAMPLE

2

19

J:l-

e 20zW..J

20

25 7 5E 11

ffi Bu

~3- -w-I-

J:e" 20

wJ:

15

2 3 4 5

SAM P L E A

E 25u-

did not permit it to be distinguished from M. ameri­canum (Hoffstetter, 1954), because the teeth were frag­mentary and morphologically similar to those of M.americanum.

The question of the validity of the epithet laurillardiresurfaced when continued discovery ofnew and oftenmore complete material showed that all adequatelyknown megatheres from Brazil belonged to Eremo­therium. During the 1940s and 1950s, when the rela­tionships and geographical distribution of Eremo­therium were still unclear, Hoffstetter (1949) and PaulaCouto (1950 , 1954, in Hoffstetter, 1949) consideredE. laurillardi to be a valid but poorly understood spe­cies. In the following paragraph Hoffstetter (1954:746­7) both supported and cast doubt on the validity oflaurillardi: "II y a done une haute probabilite pour quel'espece de Lagoa Santa soit aussi un Eremotherium etdoive recevoir Ie nom de E. laurillardi (Lund 1842).lis s'agit vraisemblablement de la meme espece que

FIGURE 3. Box and Whisker Plot of height (A) and length(B) of astragali of Eremotherium laurillardi. Horizontal lineindicates mean; vertical line indicates range; box includesone standard deviation above and below the mean; numberabove vertical line indicates observations (n) for each sample.See Appendix 1 for explanation of samples. Kruskal-Wallistests indicate that no significant differences exist among sam­ples (for height, x2 approximation = 5.56, DF = 4, P =0.2347; for length, x2 approximation = 2.65, DF = 4, P =0.6180).

South, Central, and North American Eremotheriumremains may be considered con specific.

DISCUSSION OF NOMENCLATURE

The oldest specific epithets assignable to Eremo­therium are E. laurillardi (Lund, 1842) and E. couperi(Harlan, 1842). The latter was described by Harlan(1842) as the femur ofa marine turtle Chelonia couperi,from coastal Georgia, USA (Ray, 1979). Ray (1979 :II) correctly reidentified it as a right clavicle of Ere­motherium, and stated that "it seems highly probablethat all megatheriid specimens from the late Pleisto­cene ofcoastal Georgia, including the holotype ofChe­lonia couperi , belong to a single species." Ray's hy­pothesis is justified; the size and morphology of theclavicle fall within the range of variation observed inthe Jacobina and Daytona Beach collections. Ray (1979)suggested that couperi be regarded as an unused seniorsynonym, and that mirabile (Fig. 4), though younger,be used for the North American eremothere, shouldthe two be shown to be con specific. His reasons werebased on maintaining nomenclatural stability, as mir­abile was "a name of virtually universal familiarity tovertebrate paleontologists" (Ray, 1979: 12), whereascouperi had not "entered into the literature of mega­theres at all, and into that of edentates only throughthe single mention by Hay (1923:370). Other than thefew mentions of the taxon cited above, there seems tobe no notice of it in the literature of fossil turtles"(Ray , 1979: II). While acknowledging that Ray's (1979)opinion is reasonable, we feel that couperi should beconsidered as a potentially val id name, because Harlandescribed and figured the clavicle. The remainder ofthis discussion continues largely as though couperi didnot exist, because the name has not entered into thetaxonomic history of eremotheres except for Ray's(1979) work.

Eremotherium laurillardi (Lund, 1842) is based ona molariform (ZMUC 1130) recovered from the cavesof Lagoa Santa, Minas Gerais, Brazil. The author re­ported another molariform (ZMUC 1131), and re­marked that the molariforms (Fig. 5) were approxi­mately one-quarter the size of molariforms of Me­gatherium and indicated the existence of a species thesize of a tapir. Only ZMUC 1130 was mentioned, butboth were figured by Lund (I 842:pl. 35, figs. 6, 7; re­produced in Paula Couto, 1950:pl. 35, figs. 6, 7). Ear­lier, Lund (1840) had assigned a third tooth (ofan adultindividual) from the caves to Megatherium cuvieri(=Megatherium americanum), figured by Lund (1842:pI. 36, figs. I, 2; reproduced in Paula Couto, 1950:pl.36, figs. I, 2).

Winge (1915), followed by Hoffstetter (1952 , 1954),explained that the smaller teeth showed juvenile char­acteristics and suggested that they probably belongedto the young of the species represented by the largertooth, which he assigned to M . americanum. Winge's(1915) arrangement was readily supported by subse­quent authors, as the material from Lagoa Santa in­dicated the presence of a megatheriid, but apparently

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

CARTELLE AND DE IULIIS-PANAMERICAN GROUND SLOTH

A

835

10 em

BFIGURE 4. Type specimen of Eremotherium mirabile (USNM 830) ; A, occlusal; B, lateral views of the mandible.

celie de Bahia, dont l'etude en cours nous apporteraune connaissance detaillee. II est meme possible quese soit la merne forme qui a vecu au Venezuela, de sortque E. rusconii (Schaub) pourrait tomber en synony­mie devant Ie nom de Lund, si toutefois l'on decidede prendre en consideration Ie type tres insuffisant pro­pose par Ie demier auteur."

Gazin (1957:346) believed that the two teeth com­prising the type material of E. laurillardi were "evi­dently immature and clearly inadequate for diagnosticpurposes, hence should probably be ignored or the nameM . laurillardii be left as a synonym of M . americanumas treated by Winge." The latter suggestion is techni-

cally irnpermissable, but the first has some validity, aseven isolated mature teeth ofEremotherium cannot bedistinguished consistently from those ofMegatherium.Gazin (1957) further proposed that if a single Centraland South American species were present, its validname would be E. rusconii . Paula Couto (1950) statedthat E . laurillardi was insufficiently defined, and he(1970) referred to the species from Brazil as E. lundi.Later, Paulo Couto (1978) did not refer directly to E.laurillardi, but alluded to its invalidity by agreeing withGazin that E. rusconii was valid for the Central andSouth American species. Paula Couto (1979) statedthat E. laurillardi was insufficiently defined.

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

836 JO URNAL OF VERTEBRATE PALEON TOLOGY, VOL. 15, N O. 4, 1995

FI G URE 5. E remotherium laurillardi. A, type specimen(ZMUC 1130); B, ZM UC 1131 in side view.

ditions. Thus, the migration ofMegatherium into whathas been con sidered traditionally intertropical regionsmight be expected. Ho wever, remains ofM egatheriumhave never been found, despite the reco very of variousother temperate taxa.

Th e onl y possibly valid basis for suppressing theep ithet laurillardi would be th at its type is not an ob­jective standard of reference by which the applicationof the name it bears is determined (see ICZN, Art.61a). As discussed above, various authors have alludedto thi s apparent deficiency in suggesting that the namebe ignored , and replaced by a more suitable type . It isworth considering, then, whether the mandible of E.m irabile (U SNM 830; Fig. 4), from Skidaway Island,Georgia, USA, figured by Leidy (1855 :pl. 15, figs. I ,2) adequately represents the panamerican spec ies.Hoffstetter (1952) commented that the jaw possiblyrepresented Plesiomegath erium, but nonetheless des­ignated it as the lectotype for E. mirabile. Leidy (1855)figured onl y the left mandible of this specimen, but theright also exists. Thus, the lectotype is a nearly com­plete mandible lacking angular and coronoid processes,and rostral portions of the symphyseal spout. The mo­lariforms are incomplete, largely broken at the alveolarborder; right ml is ab sent.

The jaw possesses three characters that identify it asEremotherium: the presence of a premolariform con­cavity on the jaw's lateral surface; the shallower ventralprojection of the ventra l margin of the mandible thanin Megatherium; and the more ventral position of theangular process than in Megatherium. Ho wever, thesefeatures are also present in the recently discovered,large- sized eremothere species from the Blancan ofFlorida (Hulbert et aI., 1989); the formal decription ofthis species is in progress by De Iuliis and Cartelle.Further, there exists the possibility of confusing thepanamerican species with the new species, given thegeographical proximity of the localities yielding theirremains in Florida. It is clear, on the other hand, that

8

2cm

A

Cartelle and Boh6rquez (1982), however, resurrect­ed E . laurillardi for remains from Brazil, based on newmaterial from Toea das Oncas (Jacobina), and recog­nized E . m irabile for North American remains and E .rusconii for Central and South Am erican remains ex­clus ive of Brazil (see also Cartelle and Boh6rquez,1986). Toledo (1986 , 1989) considered E. laurillardias the valid Brazilian species. Curvello and Guerin(1993) recognized E . lundi as the Brazilian species.They rejected E. laurillardi largely because its defini­tion of size did not agree with adult specimens sub­sequently reco vered.

The choice of a valid name for the panamericaneremothere is thus not simply a matter of using theoldest available name. It would seem, based on thework of Hoffstetter (1952, 1954) , Ga zin (1957), andPaula Couto (1978 , 1979) that the epithet laur illardishould be ignored. However, Cartelle and Boh6rquez(1982:48) stated that it has priority over other specificnames: " Alguns autores . . . inclinam-se para a hi­p6tese de ser 0 genero Eremotherium monoespecifico.Caso confirmado tal hip6tese, . . . , 0 nome valido paratal especie {mica seria E . laur illardi, em cuja sinonimiacairiam todos os demais nomes propostos . .." They(1982:51) j usti fied their assertion by stating that " Adiagnose feita por Lund, levando-se em conta a epo ca,parece-nos suficiente para a validade da denominacaoespecifica. " A reasonable case ma y be made in supportof th is opinion. It is true that Lund's (184 2) juvenilemolariforms cannot be distinguished from those ofMegatherium , but all subsequent diagnostic materialfrom Brazil (see e.g., Cartelle, 1992; Cunha et al ., 1985 ;Curvello and Guerin, 1993; Guerin , 1991 ; Oli veira andDamasceno, 1987 ; Paula Couto, 1975 ; Raney, 1981 ;Rolim, 1974 ; Simpson and Paula Couto, 1981; Toledo,1989) and all intertropical regions of the New Worldreported up to the present belong to Eremotherium.

The recent discovery by Cartelie of as yet unpub­lished remains reinforces the probable geographic ex­clusion ofM egatherium from this region. Over the pastfive years Cartelle has recovered, from caves in Bahia,remains that offer better understanding of the natureof faunae in intertropical Brazil. Th e caves containfaunae that preserve synchronous assemblages of typ­ically southern or temperate taxa, such as Morenela­phus sp., Myocastor coypus Molina, Glyptodon clavipesOwen , Toxodon platensis Owen; and those typicallyintertropical, such as Trigonodops lopesi Ro xo, Ho­plophorus euphractus Lund, and Xen orhinotheriumbahiense Cartelie and Lessa . The di scovery of the tem­perate Lestodon armatus (Paula Couto , 1973) in theState of Sao Paulo is additional support for this pos­sibility; this is the most northern, unambiguous recordfor this species, as that reported by Simpson and PaulaCouto (198 1) is un certain. Further , Carte lie (1992) re­ported Ocnotheriurn giganteum (Lund) from Bahia andMinas Gerais, confirming the pres ence of intertropicalLestodontinae. Such mixed assemblages make it plau­sible that temperate taxa spread northward during thelate Pleistocene, possibly due to changing climatic con-

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

CARTELLE AND DE IULIIS-PANAMERICAN GROUND SLOTH 837

the numerous remains recovered from the PleistoceneofBrazil represent the same species described by Lund(1842) .

Thus, the mandible (USNM 830) is not particularlymore adequate as an objective standard of referencethan the molariform (ZMUC 1130). Pursuing the searchfor such standards would, in our opinion, cause con­siderably more confusion, and the designation of onewould be based on somewhat subjective criteria. Weagree with Ray (1979: 11) that "Absolutely rigorousdemands of morphological adequacy for nineteenthcentury holotypes in vertebrate paleontology are ...contrary to the goal of nomenclatural stability. Oldnames may often be bolstered by invocation of geo­graphic or stratigraphic propinquity, by the additionof new . . . material." We feel that there is, therefore,no valid reason for suppressing the name Eremother­ium laurillardi, and recognize it as a potentially validname for the panamerican eremothere.

The two oldest available names are thus E . lauril­lardi (Lund, 1842) and E. couperi (Harlan, 1842). Thepublication date ofthe latter is July 6, 1842 (C. E. Ray ,pers . comm., 1994). We have been unable to determinea more precise date for E. laurillardi; the Royal DanishAcademy ofSciences and Letters and the printing houseresponsible for publication of the journal do not havearchives bearing on this matter (T. Hatting, ZMUC,pers . comm., 1994). Therefore, an objective decisionofpriority is not possible. We believe that E . laurillardishould be given preference because it has entered con­sistently into the literature of megatheriines, whereasE . couperi is virtually unknown to vertebrate pale­ontologists. We formally propose, therefore, that E .laurillardi be considered the valid name for the Pan­american, large-sized, Pleistocene eremothere. We notethat Cartelle (1992) recognized E. mirabile. At thatpoint in our collaboration, we felt that this name wouldbe most appropriate in maintaining nomenclatural sta­bility. We have since realized, clearly , that past objec­tions to E . laurillardi could not be justified.

Lund (1842) based E. laurillardi on the molariformZMUC 1130. He did not state this explicitly, but al­luded to his figure (1842 :pl. 35, fig. 6) of ZMUC 1130in a footnote; ZMUC 1131 was also figured (1842:pl.35, fig. 7). Although Lund (1842) did not formallydesignate ZMUC 1130 as the type, it is clear from histext that he based the species on ZMUC 1130. TheICZN, Art. 73a, i states that a specimen is the holotypeby original designation "Ifan author when establishinga new nominal species-group taxon states in the orig­inal publication that one specimen, and only one speci­men, is the holotype, or "type", or uses some equiv­alent expression, .. ." However, the specimen is aholotype by monotypy "If the nominal species-grouptaxon is based on a single specimen, either so statedor inferred in the original publication or demonstratedfrom evidence derived from outside the work itself.. ." (ICZN, Art. 73a, ii).

The directions of the Code do not permit an un­equivocal designation for ZMUC 1130. Ambiguity

stems from interpretation of the phrase "or someequivalent expression" (Art . 73a, i), If the phrase isintended to permit latitude with regard to the use andconcept of a type specimen during Lund's time, thenwe feel that Lund (1842) clearly intended ZMUC 1130to be regarded as a "type." Therefore, we suggest thatZMUC 1130 be considered the holotype by originaldesignation.

The type locality is Lapa Vermelha, Vale do Rio dasVelhas, Lagoa Santa, Minas Gerais, Brazil. Lund (1842)stated that ZMUC 1130 and ZMUC 1132 were fromVale do Rio das Velhas, but Lund (1843) listed themas from Lapa Vermelha (see also Paula Couto, 1950:544) . The range ofE. laurillardi (Fig. 6) extended fromSouth Carolina (Hay, 1923, reported probable remainsof this species from New Jersey), USA, to Rio Grandedo SuI, Brazil.

The following specimens from Georgia are discussedbecause of their historical signifance. Leidy (1855) de­scribed the posterior part of a cranium (USNM 832),and the left astragalus (part ofUSNM 837), which werefigured by Hodgson (1846:figs. 1, 2, 5, 6). We wereunable to locate the ungual phalanx figured in Hodgson(1846:figs. 3, 4). USNM 830, 832, and 837 are cata­logued as cotypes (in the sense ofsyntypes) ofE. " mir­abile," These specimens are from the Pleistocene ofSkidaway Island, Georgia, and were collected and do­nated to the Smithsonian (the National Institute, at thetime) by either J . P. Scriven or by Scriven and J. C.Habersham. Still other material described or knownto Leidy are USNM 825-829 (isolated molariforms)and USNM 831 (distal end ofleft humerus). The axis,cervical vertebrae, and head of a femur discussed byLeidy (1855) could not be located.

Additional material from Skidaway Island resides inANSP, including the molariform (the smaller of twocatalogued ANSP 12534) collected by Major Leconteand figured by Leidy (1855:pl. 15, fig. 4). ANSP alsohouses the E. mirabile specimens from Darien andBrunswick Canal, Georgia (collected by J. H. Couper)discussed by Leidy (1855), such as a portion of the leftdentary, ANSP 12523 . Leidy (1877:pl. 34, figs. 42, 43)also figured a molariform (ANSP 12532) from the Ash­ley Phosphate Beds, South Carolina, USA.

CONCLUSIONS

Many species have been proposed for the remainsoflarge-sized Eremotherium remains. Ofthese species,three are currently recognized in the literature: E . laur­illardi from Brazil; E. rusconii from Central Americaand the northern part of South America; and E . mir­abile from the United States. The smaller species E.elenese is poorly established. Eremotherium is mor­phologically similar to Megatherium, but may be clear­ly distinguished on differences in the skull and manus.

The species proposed for Eremotherium over thepast half-century were based largely on geographicalprovenance, without detailed comparison with estab­lished taxa. This practice resulted in a series ofspecific

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

838 JOURNAL OF VERTEBRATE PALEONTOLOGY. VOL. 15. NO .4, 1995

FIGURE 6. Map showing approximate positions of major localities yielding remains of Eremotherium laurillardi. Brazil : 1,Rio Grande do SuI; 2, Parana; 3, Mato Grosso do SuI; 4, Rio de Janeiro; 5, Espirito Santo; 6, Minas Gerais; 7, Goias; 8,Bahia; 9, Sergipe; 10, Pernambuco; 11, Paraiba; 12, Rio Grande do Norte; 13, Ceara; 14, Piaui; 15, Acre. Peru: 16, Piura.Ecuador: 17, Guayas. Panama: 18, Herrera. Colombia: 19, Huila; 20, Cundinamarca; 21, Norte de Santander. Venezuela: 22,Lara; 23, Falcon. EI Salvador: 24, San Miguel. Honduras: 25, Copan. Mexico: 26, Tabasco; 27, Jalisco; 28, Guanajuato. USA:29, Texas; 30, Florida; 31, Georgia; 32, South Carolina.

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

CARTELLE AND DE IULIIS-PANAMERICAN GROUND SLOTH 839

diagnoses based on relatively minor quantitative andqualitative differences among specimens from inade­quate samples. A wide range of variation in size andmorphology is observed in large, recently recoveredcollections of Eremotherium remains from Jacobinaand Daytona Beach. These collections demonstrate thatthe quantitative and qualitative characters cited in theliterature in support of specific distinctions are notvalid. Instead, the evidence supports the existence ofa single, Panamerican, large-sized Eremotherium spe­cies.

Eremotherium laurillardi (Lund, 1842) and E. cou­peri (Harlan, 1842) are the oldest, potentially validnames. An objective decision of priority cannot bemade, because a more precise date of publication can­not be determined for E . laurillardi. However, it hasbeen used by and is well-known to vertebrate pale­ontologists, whereas E. couperi has largely been ig­nored. Therefore, E. laurillardi is clearly more appro­priate, and recognized as the valid name for the pan­american, large-sized, Pleistocene eremothere. The typeis ZMUC 1130, a juvenile molariform. The range ofE. laurillardi extends from South Carolina (and pos­sibly New Jersey), USA, to Rio Grande do SuI, Brazil(Fig. 6). Eremotherium couperi, E . rusconii (Schaub1935), and E. mirabile (Leidy, 1855) fall as juniorsynonyms.

ACKNOWLEDGMENTS

We are grateful to Dr. S. D. Webb for his suggestionson the manuscript, and his continued support and en­couragement throughout our collaboration. We thankthe following people for allowing us to examine spec­imens in their care: K. Aaris-Serensen (ZMUC), S. A.Azevedo (MNRJ), T. Daeschler (ANSP) , R. J . Emry(USNM), J. J. Flynn (FMNH), J . Hooker (BMNH), C.McGowan (ROM), M. C. McKenna (AMNH), A. Ran­cy and J. C. Rodrigues dos Santos (UFA), D. Ste. Claire(DMAS), S. D. Webb (UF). We thank Drs . C. S.Churcher, C. E. Ray, R. Cifelli, and an anonymousreviewer for critical review of the manuscript; theircomments improved substantially the quality of themanuscript. We are grateful to H. D. Sues, G. S. Mor­gan, and K. L. Seymour for their help and patienceduring our numerous discussions on taxonomy, andA. G. Edmund for sharing with us his vast knowledgeofmega theres; D. R. Kozlovic for suggestions and helpon statistical procedures; T. Hatting for help on pub­lication dates; Robert Campochiaro, Celestino De Iuliis,Virginia Filippi, and Aleandra Reali-Rafferty for theirhelp in preparation of the manuscript; Eckard Glock­mann and Jim Dix for help with the calipers; Hum­berto Do Spirito Santo (Fig. 1) and Bruno G. Camara(Fig. 5) for the artistic work ; John Glover and PeterReali for help with photographic materials. The re­search for this paper was funded by the UniversityAffiliation Program between the University of Florida(Gainesville), the Universidade Federal de Minas Ger-

ais (Belo Horizonte), and the Smithsonian Institution,and NSERC Grant A1716 to C. S. Churcher.

LITERATURE CITED

Ameghino, F. 1889. Contribucion al conocimiento de losmamiferos fosiles de la Republica Argentina. Aetas dela Academia Nacional de Ciencias, Cordoba 6:1-1027.

Bocquentin, M. J. 1979. Mammiferes fossiles du Pleisto­cene Superieur de Muaco, Etat de Falcon, Venezuela.These doctoral, Universite de Paris, 6, 112 pp .

BUrgI, H. 1957. Artefactos paleoliticos de una tumba enGarzon, Huila. Revista Colombiana de Antropologia6:7-24.

Cartelle, C. 1992. Edentata e megamamiferos herbivorosextintos da Toea dos Ossos (Ourolandia, Bahia, Brasil).Tese de Doutorado, Universidade Federal de MinasGerais, Belo Horizonte, 30 I pp.

Cartelle, c., and G. A. Bohorquez. 1982. Eremotheriumlaurillardi Lund, 1842. Parte I. Determinacao especificae dimorfismo sexual. Iheringia, Serie Geologica 7:45­63.

Cartelle, c., and G. A. Bohorquez. 1986. Descricao daspremaxillas de Nothrotherium maquinense (Lund) Ly­dekker, 1889 (Edentata, Megalonychidae) e de Ere­motherium laurillardi (Lund) Cartelle and Bohorquez.1982 (Edentata, Megatheriidae). Iheringia, Seria Geo­logica 11:9-14.

Cunha, F. L. S., A. B. Andrade, M. H. Zucon, and M. M.Santos. 1985. Ocorrencia de mamifero fossil pleisto­cenico localizado em Monte Alegre, Sergipe , Coletaneade Trabalhos Paleontologicos. DNPM, Seria Geologica,Rio de Janeiro 27:29-34.

Curvello, M. A., and C. Guerin. 1993. L'endemisme de lamegafaune intertropicale D'Amerique du Sud as Pleis­tocene: les paresseux terrestres geants de 1'aire archeo­logique de Sao Raimundo Nonato (Piaui, Bresil). Doc­uments des Laboratoires de Geologie de Lyon 125:109­117.

De Iuliis , G ., and C. Cartelle, 1994. The medial carpal andmetacarpal elements of Eremotherium and Megather­ium (Xenarthra: Mammalia). Journal of Vertebrate Pa­leontology 14:525-533.

Edmund, A. G. 1965. A late Pleistocene fauna from theSanta Elena Peninsula, Ecuador. Royal Ontario Muse­um , Life Sciences Division 63:1-21.

Gazin, C. L. 1957. Exploration for the remains of giantground sloths in Panama. Smithsonian Report for 1956:341-354.

Guerin, C. 1991. La faune de vertebres du Pleistocene su­perieur de l'aire archeologique de Sao Raimundo Non­ato (Piaui, Bresil). Comptes rendus de l'Academic desSciences de Paris, 312, Serie 2:567-572.

Hay, O. P. 1823. The Pleistocene of North America andits vertebrated animals from the states east of the Mis­sissippi River and from the Canadian provinces east oflongitude 95°. Carnegie Institution of Washington Pub­lication 322: 1-499.

Hodgson, W. B. 1846. Memoir on the Megatherium andOther Gigantic Quadrupeds of the Coast of Georgia,with Observations on Its Geologic Features. Bartlett andWelford, New York, 47 pp .

Hoffstetter, R . 1949. Sobre los Megatheriidae del Pleisto­ceno del Ecuador, Schaubia, gen. nov. Boletin de Infor­maciones Cientificas Nacionales 3(25), suplemento:1­45 .

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

840 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 15, NO.4, 1995

1950. Rectification de nomenclature: Schaubi ­therium, nom. nov. pour Schaubia Hoffst. 1949. CompteRendu Sommaire des Seances de la Societe Geologiquede France 1950:234-235.

--- 1952. Les Mammiferes Pleistocenes de la Repub­lique de I'Equateur. Memoires de la Societe Geologiquede France, Nouvelle Serie 31(66), 391 pp.

--- 1954. Les gravigrades (Edentes, Xenarthres) descavernes de Lagoa Santa (Minas Gerais, Bresil). Annalesdes Sciences Naturelles, Zoologie, lime Serie 16:741­764 .

Hulbert, R. c., Jr., G. S. Morgan, and A. R. Poyer. 1989.Associated skeletons of megathere, pampathere, tapir,and turtles from the latest Pliocene or earliest Pleistoceneof north-central Florida. Journal of Vertebrate Paleon­tology 9, Supplement:26A.

Leidy.J. 1855. A memoir on the extinct sloth tribe ofNorthAmerica. Smithsonian Contributions to Knowledge 7:1­68.

--- 1877. Description of vertebrate remains, chieflyfrom the phosphate beds of South Carolina. Journal ofthe Academy of Natural Sciences, Philadelphia, new se­ries 8:209-261.

Lund, P. W. 1840. Nouvelles recherches sur la faune fossiledu Bresil. Annales des Sciences Naturelles. Zoologie,2ieme serie 13:310-319.

--- 1842. Blik paa Brasiliens Dyreverden fOr SidsteJordomvaeltning. Fjerde Afhandling: Fortsaettelse afPattedyrene. Det Kongelige Danske Videnskabernes Sel­skabs Skrifter. Naturvidenskabelige og MathematiskeAfhandlingar 9:137-208.

--- 1843. Blik paa Brasiliens Dyreverden fOr SidsteJordomvaeltning. Femte Afhandling. Fortsaettelse afPattedyrene. Om de nulevende og uddode Arter afRovdyrenes Familie paa det Tropiske Brasiliens IndreHoisletter. Forste Afhandling: Hundegruppen. Det Kon­gelige Danske Videnskabernes Selskabs Skrifter. Na­turvidenskabelige og Mathematiske Afhandlingar II :1­82.

Mones, A. 1973. Nota acerca de Eremotherium quanaju­atense (Duges, 1882) (Edentata, Megatherioidea) de Ar­aperos, Estado de Guanajuato, Mexico. Traballos del VCongresso Latinoamericano, Zoologia 1:162-165.

Oliveira, L. D., and J. M. Damasceno. 1987. Registro denovos especimes de mamiferos fosseis em Lagoa da Cruz­Nova-RN. Anais do Congresso Brasileiro de Paleonto­logia X, 2:189-196.

Paula Couto, C. de. 1950. Memorias sobre a paleontologiabrasileira, de P. W. Lund, revistas e comentada por C.de Paulo Couto. Instituto Nacional do Livro, Rio deJaneiro, 591 pp .

--- 1954. Megaterios intertropicais do Pleistoceno. An­ais de Academia Brasileira de Ciencias 26:447-461.

--- 1973. Edentatos fosseis de Sao Paolo. Anais daAcademia Brasileira de Ciencias 45:261-275.

--- 1970. Paleontologia da regiao de Lagoa Santa, Mi­nas Gerais, Brasil. Boletim do Museu de Historia Nat­ural da Universidade Federal de Minas Gerais, Geologia1:1-26.

--- 1975. Mamiferos fosseis do Quartenario do sudestebrasileiro. International Symposium on the Quaternary,Boletim Paranaense de Geociencias 33:90-132.

--- 1978. Marniferos fosseis do Pleistoceno do EspiritoSanto. Anais da Academia Brasileira de Ciencias, 50:365-379.

--- 1979. Tratado de Paleomastozoologia. AcademiaBrasileira de Ciencias, Rio de Janeiro, 590 pp.

Polaco-Ramos, O. J. 1981. Restos fosiles de Glossotheriumand Eremotherium (Edentata) en Mexico ; pp. 819-833in Y. T . Sanguinetti (ed.), Anais do II Congresso Latino­Americano de Paleontologia 2, Porto Alegre.

Porta, J. de. 1961. Edentata Xenarthra del pleistoceno deColombia. Universidad Industrial de Santander, Boletinde Geologia 6:5-32.

Raney, A. 1981. Marniferos fossies do Cenozoico do AltoJurua-Acre, Masters Thesis. Universidade Federal doRio Grande do SuI.

Ray, C. E. 1979. Chelonia couperi Harlan 1842, a supposedturtle based on the clavicle of a megathere (Mammalia:Edentata) Notulae Naturae of the Academy of NaturalSciences of Philadelphia 455 :1-16.

Rolim, J. L. 1974. Paleontologia e estratigrafia do Pleis­toceno Continental do Nordeste Brasileiro "FormacaoCacimbas." Masters Thesis. Universidade Federal doRio Grande do SuI.

Schaub, S. 1935. Saugertierfunde aus Venezuela und Trin­idad. Abhandlungen der Schweizerischen Palaeontolo­gischen Gesellschaft 55:1-21.

Simpson, G. G., and C. de Paula Couto. 1981. Fossil mam­mals from the Cenozoic of Acre, Brazil. III. PleistoceneEdentata Pilosa, Proboscidea, Sirenia, Perissodactyla,and Artiodactyla. Iheringia, Seria Geologica 6:11-73.

Spillmann, F. 1948. Beitrage zur Kenntnis eines neuen gra­vigraden Riesensteppentieres (Eremotherium caroli­nense, gen. et sp . nov.), seines Lebensraumes und seinerLebensweise. Palaeobiologica 8:231-279.

Stirton, R. A. and W. K. Gealey . 1949. Reconnaissancegeology and vertebrate paleontology ofEI Salvador, Cen­tral America. Bulletin ofthe Geological Society ofAmer­ica 60:1731-1753.

Toledo, P. M. de. 1986. Descricao do sincranio de Eremo­therium laurillardi Lund, 1842, taxonome, paleobio­geografia. Masters Thesis. Universidade Federal do RioGrande do suI.

--- 1989. Algumas consideracoes sobre a sistematicade Eremotherium laurillardi (Lund) CartelJe & Bohor­quez, 1982, (Edenata: Megatheriidae). Anais do XI Con­gresso Brasileiro de Paleontologia, Curitiba:763-777.

White, T. 1941. An additional record ofMegatherium fromthe Pliocene ofFlorida. Proceedings ofthe New EnglandZoological Club 19:3-6.

Winge, H. 1915. Jordfunde og nulevende Gumlere (Ed­entata) fra Lagoa Santa, Minas Geraes, Brasilien. MedUdsigt over Gumlernes indbyrdes Slaegtskab. E MuseoLundi 3:1-321.

Received 20 November 1992; accepted 21 September 1994.

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014

CARTELLE AND DE IULIIS-PANAMERICAN GROUND SLOTH 841

APPENDIX 1. Specimens and localities of astragali used in statistical analyses. Sample 3 comprises specimens from Brazilexcluding the single locality samples from Toea das Oncas, Jacobina, Bahia (Sample 1) and Pernambuco (Sample 2). L = left;R = right.

Specimen Side Localit y

Sample 1MCL 9664/02 R Toea Das Oncas , Jacobina, Bahia, BrazilMCL 9719/02 R Toea Das Oncas, Jacobina , Bahia, BrazilMCL 9733 R Toea Das Oncas, Jacobina, Bahia, BrazilMCL 9736 R Toea Das Oncas, Jacobina, Bahia, BrazilMCL 9738 R Toea Das Oncas , Jacobina, Bahia , BrazilMCL 9740 R Toea Das Oncas, Jacobina , Bahia, BrazilMCL 9744 R Toea Das Oncas, Jacobina, Bahia, BrazilMCL 9745 R Toea Das Oncas , Jacobina , Bahia, BrazilMCL 9746 R Toea Das Oncas , Jacobina, Bahia , BrazilMCL 9749 R Toca Das Oncas , Jacobina, Bahia, BrazilMCL 9752 R Toea Das Oncas, Jacobina, Bahia , BrazilMCL 9754 R Toea Das Oncas , Jacobina, Bahia, BrazilMCL 9758 R Toea Das Oncas , Jacobina, Bahia , BrazilMCL 9759 R Toea Das Oncas, Jacobina, Bahia, BrazilMCL 9760 R Toea Das Oncas, Jacobina, Bahia , BrazilMCL 9761 R Toea Das Oncas, Jacobina, Bahia , BrazilMCL 9764 R Toea Das Oncas, Jacobina, Bahia, BrazilMCL 9771 R Toea Das Oncas , Jacobina , Bahia, BrazilMCL 9772 R Toea Das Oncas , Jacobina, Bahia, BrazilMCL 9773 R Toea Das Oncas, Jacobina, Bahia, Brazil

Sample 2MNRJ 99V R Pernambuco, BrazilMNRJ 100V R Pernambuco, BrazilMNRJ 101V R Pernambuco, BrazilMNRJ 102V R Pernambuco, BrazilMNRJ 104V R Pernambuco, BrazilMNRJ 105V R Pernambuco, BrazilMNRJ 106V R Pernambuco, BrazilMNRJ 107V R Pernambuco, BrazilMNRJ 108V R Pernambuco, BrazilMNRJ 109V R Pernambuco, BrazilMNRJ 1l0V R Pernambuco, Brazil

Sample 3MNRJ417V L Conquista, Bahia, BrazilMNRJ 421V L Acre, BrazilMNRJ 279V R Paraiba, BrazilMNRJ 2130V R Boa Nova, BrazilMNRJ 2948V R Paraiba, BrazilMNRJ 2972V R Paraiba, BrazilMNRJ 3871V R Espirito Santo, BrazilBMNH M5689 L Bahia , Brazil

Sample 4ROM 22013 L Daytona Beach Bonebed, Daytona, Florida, USAROM 220 14 L Daytona Beach Bonebed, Daytona, Florida, USAROM 22015 L Daytona Beach Bonebed, Daytona, Florida, USAROM 22016 L Daytona Beach Bonebed, Daytona, Florida, USAROM 30773 L Daytona Beach Bonebed, Daytona, Florida, USA

Sample 5ROM 4036 L Coralito, Santa Elena Peninsula, EcuadorROM 4037 L Coralito, Santa Elena Peninsula, EcuadorROM 24257 L Coralito, Santa Elena Peninsula, Ecuador

Dow

nloa

ded

by [

Cas

e W

este

rn R

eser

ve U

nive

rsity

] at

11:

59 3

1 O

ctob

er 2

014